A new study published in the journal Depression and Anxiety finds a specially designed computer program can help diagnose post-traumatic stress order (PTSD) in veterans by analyzing their voices. The study also discovered that an artificial intelligence tool could distinguish with about 89 percent accuracy, between the voices of those with or without PTSD.
Charles R. Marmar, MD, the Lucius N. Littauer Professor and chair of the Department of Psychiatry at NYU School of Medicine, said that clinic might employ their findings that suggest that speech-based characteristics can be used to diagnose this disease and with further refinement and validation shortly.
At some point in their lives, more than 70 percent of adults worldwide experience a traumatic event, with up to 12 percent of people in some struggling countries suffering from PTSD. Individuals with the condition experience persistent, intense distress when reminded of the triggering effect.
Random forests, a statistical/machine learning technique was what the researchers used in the current study. The machine can "learn" how to classify individuals based on examples. The Al programs build decision rules and mathematical models that enable decision-making with increasing accuracy as the amount of training data grows.
The first recorded standard of the researchers was hours-long diagnostic interviews called Clinician-Administered PTSD Scale, or CAPS, of 53 Iraq and Afghanistan veterans with military-service related PTSD, as well as those of 78 veterans without the disease. The researchers then fed the recordings into voice software from SRI International, the Institute that also invented Siri, to yield a total of 40,526 speech-based features captured in short spurts of talk, which the team's Al program sifted through for patterns.
The random forests program linked patterns of specific voice features with PTSD, including clear speech and a lifeless, metallic tone, both of which had long been reported anecdotally as helpful in diagnosis. Though the current study did not explore the disease mechanisms behind PTSD, the theory is that traumatic events change brain circuits that process emotion and muscle tone, which affects a person's voice.
The next step for the researchers is to train the Al voice with more data to further validate it on an independent sample and apply for government approval to use the tool clinically.
The lead author, Adam Brown, Ph.D., adjunct assistant professor in the Department of Psychiatry at NYU School of Medicine said that speech is an attractive candidate for use in an automated diagnostic system, perhaps as part of a future PTSD Smartphone app, because it can be measured cheaply, remotely, and non-intrusively.
